diff options
| author | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-06-26 11:21:09 +0000 |
|---|---|---|
| committer | Simon Pilgrim <llvm-dev@redking.me.uk> | 2019-06-26 11:21:09 +0000 |
| commit | c0711af7f95ef78022362a9e7c7d562112ca9a05 (patch) | |
| tree | 9c0548ef1218177f26e8fa9b3fb67e178f8dc545 /llvm/test/CodeGen/X86/avg.ll | |
| parent | e821e79fce23d656885ceb860e31f74df948e0a4 (diff) | |
| download | bcm5719-llvm-c0711af7f95ef78022362a9e7c7d562112ca9a05.tar.gz bcm5719-llvm-c0711af7f95ef78022362a9e7c7d562112ca9a05.zip | |
[X86][AVX] combineExtractSubvector - 'little to big' extract_subvector(bitcast()) support
Ideally this needs to be a generic combine in DAGCombiner::visitEXTRACT_SUBVECTOR but there's some nasty regressions in aarch64 due to neon shuffles not handling bitcasts at all.....
llvm-svn: 364407
Diffstat (limited to 'llvm/test/CodeGen/X86/avg.ll')
| -rw-r--r-- | llvm/test/CodeGen/X86/avg.ll | 4 |
1 files changed, 2 insertions, 2 deletions
diff --git a/llvm/test/CodeGen/X86/avg.ll b/llvm/test/CodeGen/X86/avg.ll index 22a6daa999d..86d8628c9ae 100644 --- a/llvm/test/CodeGen/X86/avg.ll +++ b/llvm/test/CodeGen/X86/avg.ll @@ -382,7 +382,7 @@ define void @avg_v48i8(<48 x i8>* %a, <48 x i8>* %b) nounwind { ; AVX2-NEXT: vpshufd {{.*#+}} xmm2 = xmm0[2,3,0,1] ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm0 = xmm0[0],zero,zero,zero,xmm0[1],zero,zero,zero,xmm0[2],zero,zero,zero,xmm0[3],zero,zero,zero,xmm0[4],zero,zero,zero,xmm0[5],zero,zero,zero,xmm0[6],zero,zero,zero,xmm0[7],zero,zero,zero -; AVX2-NEXT: vpbroadcastq 24(%rdi), %ymm3 +; AVX2-NEXT: vpbroadcastq 24(%rdi), %xmm3 ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm3 = xmm3[0],zero,zero,zero,xmm3[1],zero,zero,zero,xmm3[2],zero,zero,zero,xmm3[3],zero,zero,zero,xmm3[4],zero,zero,zero,xmm3[5],zero,zero,zero,xmm3[6],zero,zero,zero,xmm3[7],zero,zero,zero ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm4 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero ; AVX2-NEXT: vpshufd {{.*#+}} xmm5 = xmm1[2,3,0,1] @@ -395,7 +395,7 @@ define void @avg_v48i8(<48 x i8>* %a, <48 x i8>* %b) nounwind { ; AVX2-NEXT: vpaddd %ymm1, %ymm2, %ymm1 ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm2 = xmm6[0],zero,zero,zero,xmm6[1],zero,zero,zero,xmm6[2],zero,zero,zero,xmm6[3],zero,zero,zero,xmm6[4],zero,zero,zero,xmm6[5],zero,zero,zero,xmm6[6],zero,zero,zero,xmm6[7],zero,zero,zero ; AVX2-NEXT: vpaddd %ymm2, %ymm0, %ymm0 -; AVX2-NEXT: vpbroadcastq 24(%rsi), %ymm2 +; AVX2-NEXT: vpbroadcastq 24(%rsi), %xmm2 ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm2 = xmm2[0],zero,zero,zero,xmm2[1],zero,zero,zero,xmm2[2],zero,zero,zero,xmm2[3],zero,zero,zero,xmm2[4],zero,zero,zero,xmm2[5],zero,zero,zero,xmm2[6],zero,zero,zero,xmm2[7],zero,zero,zero ; AVX2-NEXT: vpaddd %ymm2, %ymm3, %ymm2 ; AVX2-NEXT: vpmovzxbd {{.*#+}} ymm3 = mem[0],zero,zero,zero,mem[1],zero,zero,zero,mem[2],zero,zero,zero,mem[3],zero,zero,zero,mem[4],zero,zero,zero,mem[5],zero,zero,zero,mem[6],zero,zero,zero,mem[7],zero,zero,zero |
